Computational and quantitative methods are now an essential part of biological research. There is an urgent need for scientists who are expert in both computational and life sciences. We propose to continue addressing this need through the Columbia University predoctoral Training Program in Computational Biology. Our main goal is to train young scientists to do pioneering and high-impact research in biology using computational methods. We will accomplish this goal by providing the necessary background in both life and computational sciences and through mentored student research projects. The students will be trained to be experts in one area of computational biology (e.g. systems biology), have a good knowledge of another area (e.g. structural biology), and have an in-depth understanding of experimental biology in the area of their specialization (e.g. genetics or neuroscience). In the initial funding period - as suggested by the NIGMS training program guidelines - we focused on course and curriculum development using a small group of highly qualified trainees with diverse scientific backgrounds. Students trained by our program have already made exciting scientific breakthroughs and published high-profile articles. In this application we propose to expand the developed program to accommodate a large number of qualified students at Columbia and request 8 slots for the program. The program coursework will be tailored to students' backgrounds and interests, and will consist of a 4x2 scheme: at least 2 courses in life sciences (e.g. biochemistry, cell biology), 2 courses in quantitative subjects (e.g. machine learning, statistics), 2 courses in computational biology (e.g. computational systems biology, biophysics), and 2 electives from any of the above areas. Students will be able to earn doctoral degrees in the C2B2 Graduate Program or in the 13 affiliated departments and programs and in the laboratories of 20 training program faculty, representing both quantitative and life science disciplines. The students will also learn from rotations (including ones in experimental labs), guided independent study, an oral qualification exam, seminars, retreats, and journal clubs. The program will be directed by Dr. Barry Honig, Director of the Columbia University Center for Computational Biology and Bioinformatics (C2B2), and crucial program decisions will be made by the program's Executive Committee. It takes on average 5 years to complete our Ph.D. program and we will typically support computational biology students for a 2-3 year period. We will make every effort to recruit and retain students who belong to minority groups, come from underprivileged backgrounds, or are disabled. We will continue to accomplish this goal by actively recruiting qualified students from these groups. All students will be required to take a 1 semester course in responsible conduct of research.